bioRxiv Science⌕ Search

Biology subjects

Tsongalis, G. J.

Publications and source records attributed to Tsongalis, G. J..

2 recordsLinked to original sources

A rapid, multiplex digital PCR assay for EGFR, KRAS, BRAF, ERBB2 variants and ALK, RET, ROS1, NTRK1 gene fusions in non-small cell lung cancer

Digital PCR (dPCR) is emerging as an ideal platform for the detection and tracking of genomic variants in cancer due to its high sensitivity and simple workflow. The growing number of clinically-actionable cancer biomarkers creates a need for fast, accessible methods that allow for dense information content and high accuracy. Here, we describe a proof-of-concept amplitude modulation based multiplex dPCR assay capable of detecting 12 single nucleotide and indel variants in EGFR, KRAS, BRAF, and ERBB2, 14 gene fusions in ALK, RET, ROS1, NTRK1, and MET exon 14 skipping present in non-small cell lung cancer (NSCLC). We also demonstrate the use of multi-spectral target signal encoding to improve the specificity of variant detection by reducing background noise up to 11-fold. The assay reported an overall 100% PPA and 98.5% NPA compared to a sequencing-based assay in a cohort of 62 human FFPE samples. In addition, the dPCR assay rescued actionable information in 10 samples that failed to sequence, highlighting the utility of a multiplexed digital assay as a potential reflex solution for challenging NSCLC samples.

cancer biology↗

Mixed Effects Machine Learning Models for Colon Cancer Metastasis Prediction using Spatially Localized Immuno-Oncology Markers

Spatially resolved characterization of the transcriptome and proteome promises to provide further clarity on cancer pathogenesis and etiology, which may inform future clinical practice through classifier development for clinical outcomes. However, batch effects may potentially obscure the ability of machine learning methods to derive complex associations within spatial omics data. Profiling thirty-five stage three colon cancer patients using the GeoMX Digital Spatial Profiler, we found that mixed-effects machine learning (MEML) methods{dagger} may provide utility for overcoming significant batch effects to communicate key and complex disease associations from spatial information. These results point to further exploration and application of MEML methods within the spatial omics algorithm development life cycle for clinical deployment.

pathology↗